© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 63889
    Prev
    Next
    INDEX
    Explanations

    javascript package imports

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    日を
    -1.11
     masculina
    -0.96
     at
    -0.96
     despite
    -0.94
    日が
    -0.89
     pneu
    -0.89
    体を
    -0.87
    榊
    -0.86
    戰鬥
    -0.84
     otor
    -0.84
    POSITIVE LOGITS
     queste
    1.05
     asimismo
    0.94
    führen
    0.93
     manchen
    0.93
     ENTIRE
    0.90
    during
    0.90
     ejerc
    0.87
    kehren
    0.87
     arrogant
    0.86
     Hinsicht
    0.85
    Activations Density 0.004%

    No Known Activations